Note: Descriptions are shown in the official language in which they were submitted.
CA 02224460 2000-11-08
1
PROCESS FOR PRODUCING GM2 SPECIFIC
ANTIBODIES
FIELD OF THE INVENTION
This invention relates to the making of antibodies
against glycolipids, and the uses of these antibodies. More
particularly, it relates to processes for making antibodies
specific to the monosialogangliosides known as "GM2" and
"GM1", as well as the uses of these antibodies. The uses
include diagnostic and screening applications, as well as
l0 therapeutic modalities.
$ACKGROUND AND PRIOR ART
Gangliosides are a class of molecules which are
glycolipids. Different gangliosides have been identified as
prominent cell surface constituents of various transformed
cells, including melanoma, as well as other tumors of
neuroectodermal origin. See, e.g., Ritter and Livingston, et
al., Sem. Canc. Biol. 2: 401-409 (1991). Oettgen, VCH Verlags
Gesellschaft (Weinheim Germany 1989),
Gangliosides are known as mono-, di-, tri or
polysialoganglio- sides, depending upon the degree of
glycosylation with sialic acid residues. Abbreviations
employed to identify these molecules include "GM1", "GD3",
"GT1", etc., with the "G" standing for ganglioside, "M", "D"
or "T", etc. referring to the number of sialic acid residues,
and the number or number plus letter ( a . g . , "GTla" ) , ref erring
to the binding pattern observed for the molecule. See
Lehninger, Biochemistry, pg. 294-296 (Worth Publishers, 1981);
Wiegandt, in Glycolipids: New Comprehensive Biochemistry
(Neuberger et al., ed., Elsevier, 1985), pp. 199-260.
The monosialoganglioside GM2 has the structure:
GaINAcBI -~ 4Ga1B1 -~ 4GlcB1 ~ Ceramide
3
2aNeuAc
while GM1 has structure
CA 02224460 2000-11-08
2
Ga11~1 - 3GalNAcf31 - 4Ga1f31 - 4Glcf~l~Ceramide
3
T
2aNeuAc
As noted, supra, the gangliosides are prevalent cell surface
markers on transformed cells, such as melanoma. This has made
them attractive targets for cancer research. Livingston, et
al., Proc. Natl. Acad. Sci. USA 84: 2911-2915 (1987)
describe results of a vaccine
based trial, wherein subjects afflicted with melanoma
received, as vaccines, either whole cells which present high
levels of GM2, pure GM2 or pure GM2 plus bacterial adjuvant.
Attention is also drawn to Livingston, et al., J. Clin. Oncol.
12(5): 1036-1044 (1994), and Irie, et al., U.S. Patent No.
4,557,931, and
deal with the use of GM2 as a vaccine.
There has also been interest in the manufacture and use
of antibodies, such as monoclonal antibodies, which bind to
gangliosides. Such antibodies share the common
characteristics of all antibodies, including antigenic
specificity, and are of interest both diagnostically and
therapeutically. See, e.g., Kjeldsen, et al., U.S. Patent No.
5,229,289: Nudelmann, et al., U.S. Patent No. 5,240,833:
Hakomori, et al., U.S. Patent No. 5,308,614, and Hakomori, et
al., U.S. Patent No. 5,389,530. All of these patents
deal in large part with the
generation of ganglioside specific antibodies.
There are difficulties unique to the immunology of
gangliosides, which are touched upon briefly here. First,
while these molecules are prevalent on transformed cells, they
are also common on certain normal cells, such as neural cells.
There is a risk, in administering gangliosides to a subject,
that the resulting antibody response will damage normal cells.
Indeed, certain autoimmune pathologies, such as Guillain-
Barre' Syndrome, are characterized by autoimmune antibodies
reactive with GMl or GQib. See, e.g., Yuki, et al., J. Exp.
CA 02224460 2000-11-08
3
Med. 178: 1'1-71-1775 (1993); Aspinall, et al., Infect & Immun.
62(5): 2122-2125 (1994).
There is an additional practical problem in that
gangliosides are extremely difficult to secure in amounts
sufficient for immunization protocols. No practical synthetic
method is presently available. As a result, gangliosides are
secured via purification from tissue, such as bovine cranial
tissues. Even under optimum conditions, the yields of pure
gangliosides are vanishingly small. Further, purification
from mammalian tissue carries with it the risk of transmitting
contaminants such as viruses, prion particles, and so forth.
Alternate methodologies for securing ganglioside specific
antibodies are thus highly desirable.
Lipopolysaccharide, or "LPS" molecules are found on the
surface of Gram negative bacteria. Many of these molecules
are quite toxic, leading to toxic shock syndrome, endotoxemia,
and other conditions. There is a great deal of diversity in
the LPS molecules of different bacteria. Indeed, even within
the confines of a particular type of bacteria, the LPS
molecule may differ, between various serotypes/serovars.
The LPS molecules of Cam~,vlobacter jej-uni ("C. jejuni")
have been studied in some detail. Representative, but by no
means exhaustive, of the studies on these LPS molecules, are
Aspinall, et al., Eur. J. Biochem. 213: 1017-1027 (1993):
Aspinall, et al., Biochem. 33: 241-249 (1994); Yuki, et al.,
Infect. & Immun. 62(5): 2101-2103 (1994); Aspinall, et al.,
Infect. & Immun. 62(5): 2122-2125 (1994).
The Aspinall 1993
paper is of particular interest in that it presents LPS
structures for C. jejuni serovars 0:1, 0:4, 0:23, and 0:36.
Aspinall et al., state that OS: l, OS:23 and OS:36 are mutally
indistinguishable, and have chain termini identical to those
of GM2, while the OS:4 chain terminus is identical to GDia.
Some of these papers discuss certain similarities between
portions of the 0:19 LPS molecule and GM1, an idea also
discussed by Wirguin, et al., Ann. Neurol. 35(6): 698-703
(1994), as well as by Aspinall, et
CA 02224460 2000-11-08
4
al. , Infect. & Immun. 52 (5) : 2122-2125 (1994) ; Yuki, et al. ,
Infect. & Immun. 62 (5) : 2102-2103 (1994) ; Yuki, et al. , J. Exp.
Med. 178: 1771-1775 (1993). C. ieiuni bacteria are themselves
implicated in gastrointestinal disorders, and there has been
some interest and activity in developing diagnostic tests for
determining the presence of C. je~uni in samples. See PCT
Application W086/01808 to Wright et al., U.S. Patent No.
5,200,344 to Blaser, et al., U.S. Patent No. 5,169,757, to
Yamazaki, et al., and Japanese Patent Application 63-273497 on
this point. One notes that in all of these references, the
manner in which the antibodies are generated involves the use
of whole cells. Blaser, et al., speak of an antigen unique to
C. jejuni, but it is a protein, not a glycolipid.
There has been no linking of C.~ e'L LPS antigens as a
source for producing antibodies against GM2. Further, there has
been no suggestion that the antibodies produced in this way may
be used in diagnostic and therapeutic approaches to
pathological conditions, such as cancer.
The invention provides a method for making an antibody
which binds to monosialoganglioside GM2, comprising:
(i) immunizing a subject animal with an antigen comprising
the structure:
GalNAc R l~4Ga1 ( I I 3NeuAc ) Hex
in an amount sufficient to provoke production of antibody by
said animal with the proviso that the antigen is not GM2, and
(ii) isolating the antibodies produced by said animal.
In preferred embodiments the antigen is a
lipopolysaccharide, particularly a purified lipopolysaccharide
from a gram negative bacterium, more particularly Campylobacter
jeiuni, especially of serotypes 0:1 and 0:36.
The invention also provides a method for determining the
presence of monosialoganglioside GM2 in a sample, comprising
contacting the sample with an antibody produced in accordance
with the invention and determining binding of the antibody as
a determination of GM2 in the sample. The sample may contain
transformed cells such as melanoma cells, neuroblastoma cells,
sarcoma cells, glioma cells or seminoma cells.
CA 02224460 2000-11-08
4a
These are all features of the invention described herein,
in detail, in the disclosure which follows.
BRIEF DESCRIPTION OF THE FIGURES
Figure 1 is a dot blot showing results obtained, testing
various ganglioside species, with antiserum obtained following
immunization of rabbits with LPS prepared from various
Camgvlobacter jejuni serovars. Lanes 1 and 2 used antiserum
following immunization with 0:1, lanes 3 and 4 used 0:19, lanes
5 and 6 used 0:23, lanes 7 and 8 used 0:36, and lanes 9 and 10
used bovine brain ganglioside GM2. Odd numbered lanes are tests
with pre-immune serum, and even numbers are immune serum.
Figures 2A and 2B present data obtained in immune thin
layer chromatography tests. Figure 2A shows results obtained
by testing pre-immune serum. The pre-immune serum was from a
rabbit prior to immunization, tested with LPS 0:2, diluted
1:100 in PBS. The TLC was developed in two solvents (i.e.,
CH3C1 and acetone, at 1:1 ratio is the first solvent, the
second solvent is CHC13, methanol, and 0.2o CaCl2, at 55:45:10
CA 02224460 1997-12-11
WO 97/00445 PCT/US96I09824
ratio) prior to overlaying with serum. Moving across the
lanes, lane 1 shows a test with mixed bovine brain
gangliosides (10 u1) lane 2 tested bovine brain ganglioside
GM2 (1 ug), lanes 3-5 tested, respectively, human melanoma (5
~ 5 u1), human neuroblastoma (4 u1), and human renal carcinoma (5
u1) cell lines, while lane 6 tested fresh human melanoma (3
u1).
In figure 2B, immune serum from the same subject animal
was tested. The serum was taken from the fifth bleed, and
diluted 1:500 in PBS, as were the materials tested. Note that
the upper bands, in lanes 5 and 6, were also found in these
bands when pre-immune serum was tested, and thus are not a
result of the immunization. In contrast, the lower bands in
lanes 5 and 6 are the result of the immunization, while the
upper bands are not.
Figures 3A and 3B depict the results obtained in antibody
dependent cellular cytotoxicity tests. The serum used was
obtained following immunization with LPS serotype 0:1. In
figure 3A, renal cell carcinoma cell line SK-RC-9, which is
known to present GM2 on its surface was tested, while in
figure 3B, colon carcinoma cell line SW 1222, which does not
present GM2, was tested. In both figures, the open circles
present data points where effector cells, target cells, and
immune serum were combined. Filled in circles present data
points from tests involving effector cells, target cells, and
pre-immune serum, while filled in squares represent the
combination of target cells and immune serum without effector
cells.
DETAILED DESCRIPTION OF PREF~RFD E1~30DIMENTS
Exam; a 1
All bacteria used were obtained from the American Type
Culture Collection. Samples of Camnvlobacter jeiuni ("C.
jejuni" hereafter) of serotype 0:1, 0:19, 0:23, and 0:36 were
used. For each sample, the lipopolysaccharide antigens ("LPS"
antigens hereafter) were extracted from heat inactivated
bacteria, using the hot phenol-water methodology of Westphal,
CA 02224460 2000-11-08
6
et all , Z . Naturforschlag. 7b: 148-155 ( 1952 ) ,
with minor modifications. Specifically, RNA
was removed by digestion with commercially available bovine
pancreas RNA, followed by size exclusion chromatography on
Sephade5~ G-25 in accordance with Goldman, Meth. Enzymol. 138:
267-275 (1987). The extracted LPS was lyophilized, and stored
at -20°C until ready for use.
Example 2
The LPS preparations were then used to prepare
immunogenic compositions. In each case, the LPS antigen was
mixed with adjuvant. For each composition, two rabbits were
used as subject animals (female New Zealand white rabbits, 2
2.5 kg in weight). On day 0, the animals were injected,
intradermally, with 500 ug of LPS in Complete Freunds
Adjuvant. Fourteen days later, the animals received a second
injection of 500 ug of LPS, this time in Incomplete Freund's
Adjuvant. This was followed up with injections at day 28 and
day 42, in each case with 250 ug of LPS in Incomplete Freund's
adjuvant. Two animals were injected with pure bovine brain
GM2, (200 ug, 200 ug, 100 ug, 100 ug, at the same time points
in the experiment). All animals were bled at days 0, 14, 28,
42 and 64, to determine antibody titer and to conduct
additional analyses. Some rabbits were also given booster
injections of 250 ug LPS in IFA, 13 and 17 weeks after the
fourth injection, and were bled two weeks after the last
booster injection.
Example 3
With reference to the immunization protocol described,
supra, following each bleed, sera from the immunized rabbits
were analyzed, via a standard ELISA, to determine (i) if
antibodies against GM2 were present, and (ii), if so, were
these of class IgG, or IgM. This was done using commercially
available antirabbit IgM or IgG antibodies. Immunization with
_C. jeiuni serovar 0:1 yielded IgG antibodies, and high titer
serum ( 1:3200, and >_ 60, 000 after booster injections) , as well
as low titer IgM antibodies (1:100), both of which'recognized
GM2. The immunization with ~. ~eiuni 0:36 yielded moderate
*Trade-mark
CA 02224460 1997-12-11
WO 97/00445 PCT/US96/09824
7
titer IgM antibodies (1:800), and low titer IgG antibodies
(1:100). These also recognized GM2. The immunization with
0:19 yielded high titer IgG antibodies (peak: 1:12800), and
moderate to high titer IgM antibodies ( peak: 1:1200 ) : however,
- 5 these antibodies recognized ganglioside GM1. The test with
pure, bovine brain GM2 yielded low titer IgM (1:200), and IgG
(1:200).
Example 4
The sera obtained as described, supra, was analyzed
thoroughly for specificity in dot blot assays. The sera were
stained for IgM and IgG antibodies to GM2, and also to a panel
of other gangliosides derived from bovine brain material,
including GM3, GM1, GD3, GD2, GDla, GDlb, GTlb, and GQlb,
immobilized on nitrocellulose paper. The results follow, in
Table 1. The dot blot assay was a standard one, using serum
dilutions of 1:100 and commercially available anti-rabbit IgG
and IgM antibodies. In table 2, which follows, a grade of
"3+" means that there was strong staining, "2+" a medium-
strong stain, "1+", a weak stain, and "+/", a trace of
staining. Figure 1, which follows, presents the dot blots.
CA 02224460 1997-12-11
WO 97/00445 PCT/US96/09824
8
TABLE 1
IgG reactivity of immune sera with gangliosides and LPS as .
determined by dot blot
immune
stains
Animal Vaccine Serum* Ganglioside
GM3 GM2 GM1 GD3 GD2 GDla GDlb GTlb
l0 1 LPS 0:1 pre-vacc. - - - - -
-
1 " 4. vacc. 3+ - - - _ _ _
-
2 " pre-vacc. - - - - - - -
-
2 " 4. vacc. 2+ - _ _ _ _ _
-
3 LPS0:19 pre-vacc. - - - - - -
- 1+ 2+ -
3 " 4. vacc. - 3+ - -
- -
" pre-vacc. - - - - - -
- - 3+ -
4. vacc. - 3+ - -
-
5 LPS0:23 pre-vacc. - - - - - - -
-
5 " 4. vacc. _ _ _ _ _ _ _
-
6 " pre-vacc. - - - - - - -
-
6 " 4. vacc. _ _ _ _ _ _ _
-
7 LPS0:36 pre-vacc. - - - - - - -
-
7 " 4. vacc. - - - - - - -
-
g " pre-vacc. - - - - ' - -
-
_
8 " 4-vacc. - - - - - -
-
9 GM2 pre-vacc. - - - - - - -
-
g " 4. vacc. - 1+ - - - - - -
10 " pre-vacc. - - - - - ' ' -
10 " 4. vacc. - 2+ - _ _ _ _ _
*serum dilution 1:100
As will be seen in Table 1, there was strong serum IgG
reactivity with GM2 -from animals immunized with C. jejuni
serovar 0:1 LPS, while the other serum samples tested
negative. The sera from C_. j_e'Zuni serovar 0:19 immunized
animals showed strong reactivity with GM1, and moderately
strong activity with GDlb. Weak activity with GDla was also
seen.
Example 5
In further experiments, the sera were tested, in immune
thin layer chromatography, for reactivity with purified bovine
brain GM2, and with GM2 extracted from fresh human melanoma,
a human renal cell carcinoma cell line, a human melanoma cell
line, and from a human neuroblastoma cell line. The immune
CA 02224460 1997-12-11
WO 97/00445 PCT/US96/09824
9
thin layer chromatography tests were carried out using
standard methodologies. The IgG antibodies obtained from sera
of animals immunized with C. ieiuni serovar 0:1 reacted
strongly with all sources of GM2. Figure 2A and 2B show these
results. The IgG antibodies obtained from sera obtained from
C. iejuni serovar 0:19 immunized animals reacted strongly with
purified bovine brain GM1.
Example 6
The ability of the GM2 specific sera to bind to cell
surfaces was tested in a standard mixed hemadsorption assay.
In this assay, human renal carcinoma cell line SK-RC-9, which
expresses high levels of cell surface GM2 on its surface, was
combined with the sera obtained from animals immunized with C.
_ieiuni serovar 0:1 LPS. the antibodies in the sera were
surface reactive (peak titer 1:600). See following Table 2.
CA 02224460 1997-12-11
WO 97/00445 PCT/US96/09824
TABLE 2
Cell surface reactivity of immune sera with SK-RC-9 as .
determined by mixed hemasorption assays
5
Animal Vaccine Serum Titer*
1gM 1gG
1 LPS C. jejuni 0:1 pre-vacc. - +/-
l0 1 " 2. wks 1. vacc. +/- 1:100
post
1 " 2. wks 2. vacc. - 1:300
post
1 " 2. wks 4. vacc. - 1:600
post
2 " pre-vacc. - +/-
2 " 2. wks 1. vacc. 1:100 +/-
post
2 " 2. wks 2. vacc. - +/-
post
2 " 2. wks 4. vacc. +/- 1:150
post
2 " 2 wks. 6. vacc. 1:2000 1:1600
post
2 9 GM2 ganglioside pre-vacc. +/- -
o
g " 2. wks 1. vacc. +/- -
post
9 " 2. wks 2. vacc +/- -
post
g " 2. wks 4. vacc 1:300 1:400
post
2 10 " pre-vacc. - +/-
5
10 " 2. wks 1. vacc. +/- +/-
post
10 " 2. wks 2. vacc. 1:100 +/-
post
10 " 2. wks 4. vacc. +/- +/-
post
30 * Starting dilution of 1:100
serum:
Example 7
Further experiments were carried out to test for the
35 presence of antibodies which mediate antibody-dependent
cellular cytotoxicity ("ADCC"). These experiments involved
the well known 5lCr release assay.
CA 02224460 1997-12-11
WO 97/00445 PCT/US96/09824
11
In these experiments, serum obtained from rabbits which
had received six vaccinations with LPS isolated from C. ~eiuni
~ serotype 0:1 was combined with various tumor cell lines. The
antibodies were cytotoxic to GM2 expressing tumor cell lines,
- 5 including renal cell carcinoma cell line SK-RC-9 (up to 60%
specific killing, at serum dilutions of 1:800), and
neuroblastoma cell line 1MR-32 (up to 80o specific killing, at
serum dilutions of 1: 800 ) : SK-Mel-31 ( a melanoma cell line: up
to 35 o specific lysis) , and renal cell carcinoma SK-RC-49 ( 35%
to specific lysis, at 1:400 serum dilutions. There was no
specific lysis found for those cell lines which did not
present GM2 on their surfaces. Normal rabbit serum did not
mediate cell lysis either. This will be seen in figure 3.
The foregoing examples set forth a process for making
15 antibodies specific for a ganglioside, i.e., GM2 or GM1. The
methodology is surprising in that the immunogen is not a
ganglioside. As is seen, su r , immunization with a molecule
of structure GaINAcBI-4Ga1(II'NeuAc)-Hex, generated anti-GM2
specific antibodies, while immunization with an immunogen of
20 structure GalB1-3GaINAcBI-4Ga1(II'NeuAc) generated anti-GM1
specific antibodies. These molecules are not gangliosides.
The immunogen used is preferably a lipopolysaccharide
molecule, most preferably an LPS -molecule secured from C.
_ieiuni. As will be seen from the discussion, su ra, different
25 serovars of C. jeiuni have different LPS molecules on their
surface. It is well within the skill of the artisan to carry
out analyses to determine if a molecule, such as an LPS
molecule has the desired structure as per Aspinall, et al.,
1993, supra, e.g. Thus, the invention is not limited to the
30 use of the specific serovars discussed herein.
It is preferred that the immunogens be administered
together with an adjuvant, such as the described Complete and
Incomplete Freund's Adjuvant. Other adjuvants, including,
e.g., Bortadella ~ertussis, aluminum hydroxide, QS-21, BCG,
35 Salmonella minnesota 8595, adjuvant, MPL and other adjuvants
well known to the art may be used. Similarly, the immunogen
may be used "as is", such as described herein, or it may be
CA 02224460 1997-12-11
WO 97/00445 PCTIUS96109824
12
coupled to materials which enhance its immunogenicity, such as
Keyhole limpet hemocyanin, bovine serum albumin, etc.
The subject animal may be any of the standard mammalian ,
hosts used in the art for generating antibodies, including
mice, rats, and other rodents; sheep, goats, and other .
ruminants, and so forth.
The antibodies described herein are polyclonal in nature.
It is assumed, however, that one of ordinary skill is fully
able to work with materials which generate the disclosed
polyclonal antibodies so as to develop hybridomas and
monoclonal antibodies. Such methodologies, including the well
known Kohler-Milstein method and other ways to "immortalize"
antibody producing B cells, need not be set forth here.
While one does not use humans as test animals, the
ability to generate antibodies against these LPS structures
does provide the artisan with an appropriate way to vaccinate
individuals to achieve active, rather than passive
immunization. The art is fully familiar with methodologies
which eliminate the toxic effect of LPS molecules without
significantly modifying their immunogenicity. It has been
shown, that mammalian immune systems can, and do react to
these LPS molecules to produce antibodies against GM2. It has
also been shown that various types of cancer, such as renal
carcinoma, neuroblastoma carcinoma, sarcoma, glioma, seminoma
and melanoma, are characterized in part by expression of GM2
molecules on their surface. Vaccination with non-toxic forms
of LPS, including those used in the examples described herein,
should lead to immunological clearance of GM2 presenting
cancer cells. Hence one aspect of the invention relates to a
method for treating a subject with a pathological condition,
such as a cancer, where abnormal cells present GM2 on their
surfaces, by administering an effective amount of a non-toxic
LPS molecule to the subject, so as to provoke production of
anti-GM2 antibodies in said subject.
As is shown in the examples, supra, antibodies produced
in the manner described are useful diagnostically, as in the
determination of whether or not cells are present in a sample
CA 02224460 1997-12-11
WO 97/00445 PCT/US96/09824
13
which present gangliosides GM1 or GM2 on their surface. Thus,
one is able to screen a sample for irregularities, such as the
. onset of tumorigenicity, where tumor cells present GM1 or GM2
on their surface, while non-transformed cells do not.
. 5 Melanoma and renal carcinoma are examples of such transformed
cells. Similarly, samples can be analyzed generally, using
the antibodies described herein. For example, a bovine brain
fraction can be assayed to determine whether or not desired
gangliosides GM1 and/or GM2 are present, and if so, whether or
not they are present in an amount to warrant further analysis
and purification.
The work described supra as it relates to vaccines also
provides support for another aspect of the invention, which is
a method for treating a pathological condition, such as
melanoma, neuroblastoma, sarcoma, glioma, seminoma, and other
types of cancer wherein the pathology is characterized by
appearance of GM2 or GM1 on abnormal cells. One may
administer a therapeutically useful dose of the antibodies of
the invention sufficient to provide a therapeutically positive
effect to the subject so treated.
Other aspects of the invention will be clear to the
skilled artisan and need not be repeated here.
The terms and expressions which have been employed are
used as terms of description and not of limitation, and there
is no intention in the use of such terms and expressions of
excluding any equivalents of the features shown and described
or portions thereof, it being recognized that various
modifications are possible within the scope of the invention.
